0
Research Papers: Gas Turbines: Turbomachinery

Analysis Versus Synthesis for Trending of Gas-Path Measurement Time Series

[+] Author and Article Information
S. Borguet

Turbomachinery Group,
University of Liège,
Campus du Sart-Tilman, B52/3,
Liège 4000, Belgium
e-mail: s.borguet@ulg.ac.be

O. Léonard

Turbomachinery Group,
University of Liège,
Campus du Sart-Tilman, B52/3,
Liège 4000, Belgium
e-mail: o.leonard@ulg.ac.be

P. Dewallef

Laboratory of Thermodynamics,
University of Liège,
Campus du Sart-Tilman, B49,
Liège 4000, Belgium
e-mail: p.dewallef@ulg.ac.be

These terms will be used as synonyms here despite the slight difference in their actual definition.

Contributed by the Turbomachinery Committee of ASME for publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received July 14, 2014; final manuscript received July 18, 2014; published online September 16, 2014. Editor: David Wisler.

J. Eng. Gas Turbines Power 137(2), 022603 (Sep 16, 2014) (8 pages) Paper No: GTP-14-1389; doi: 10.1115/1.4028385 History: Received July 14, 2014; Revised July 18, 2014

Gas-path measurements used to assess the health condition of an engine are corrupted by noise. Generally, a data cleaning step occurs before proceeding with fault detection and isolation. Classical linear filters such as the EWMA filter are traditionally used for noise removal. Unfortunately, these low-pass filters distort trend shifts indicative of faults, which increases the detection delay. The present paper investigates two new approaches to nonlinear filtering of time series. On the one hand, the synthesis approach reconstructs the signal as a combination of elementary signals chosen from a predefined library. On the other hand, the analysis approach imposes a constraint on the shape of the signal (e.g., piecewise constant). Both approaches incorporate prior information about the signal in a different way, but they lead to trend filters that are very capable at noise removal while preserving at the same time sharp edges in the signal. This is highlighted through the comparison with a classical linear filter on a batch of synthetic data representative of typical engine fault profiles.

FIGURES IN THIS ARTICLE
<>
Copyright © 2015 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

A notional representation of the trending problem

Grahic Jump Location
Fig. 2

A piecewise constant signal (top panel) and its first-order derivative (bottom panel)

Grahic Jump Location
Fig. 3

A piecewise linear signal (top panel) and its second-order derivative (bottom panel)

Grahic Jump Location
Fig. 4

The root signal (top panel) and the combination of atoms leading to the root signal (bottom panel)

Grahic Jump Location
Fig. 5

Graphical representation of the library used for signal reconstruction

Grahic Jump Location
Fig. 6

The test signals—jump (top panel) and ramp (bottom panel)

Grahic Jump Location
Fig. 7

Noise reduction factor of the different filters at different SNRs

Grahic Jump Location
Fig. 8

Comparison of the behavior of the different filters on a jump signal—SNR = 3

Grahic Jump Location
Fig. 9

Comparison of the behavior of the different filters on a ramp signal—SNR = 3

Grahic Jump Location
Fig. 10

First- (black dots) and second- (gray dots) order derivatives of the estimated ramp signal

Grahic Jump Location
Fig. 11

Weights on the atoms of the jump (black dots) and ramp (gray dots) library for the estimated ramp signal

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In